novoads-api

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill follows established best practices for handling API keys.
  • [DATA_EXFILTRATION]: Network requests are directed to the official vendor domain (api.novoads.ai). The skill correctly instructs users to store secrets in a .env file and warns against exposing them in logs or chat.
  • [COMMAND_EXECUTION]: The skill uses legitimate system utilities such as ffmpeg, ffprobe, and curl. These tools are used for their intended purposes: extracting metadata, processing frames, and interacting with the REST API.
  • [EXTERNAL_DOWNLOADS]: Mentions dependencies on trusted tools like Homebrew (ffmpeg) and OpenAI's Whisper (openai-whisper). These are well-known services and do not escalate the risk profile.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided media files (video/images) for analysis and cloning. This involves an inherent indirect prompt injection surface; however, the skill provides extensive documentation for output verification (Video QA) to detect and mitigate potential anomalies.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 06:16 PM
Security Audit — agent-trust-hub — novoads-api